A Yellow Polariton Condensate in a Dye filled Microcavity
Tamsin Cookson1+, Kyriacos Georgiou2+, Anton Zasedatelev3,
Richard T. Grant2, Tersilla Virgili4, Marco Cavazzini4, Francesco
Galeotti4, Caspar Clark5, Natalia G. Berloff3,6, David G. Lidzey2*
and Pavlos G. Lagoudakis1,3*
1. Department of Physics and Astronomy, The University of
Southampton, University Road, Southampton SO17 1BJ, U.K.
2. Department of Physics and Astronomy, The University of
Sheffield, Hicks Building, Hounsfield Road, Sheffield S3 7RH,
U.K.
3. Skolkovo Institute of Science and Technology, Moscow 143026,
Russia.
4. IFN, ISMAC and ISTM – CNR Milano, Italy.
5. Helia Photonics Ltd, 2 Rosebank Rd, Livingston EH54 7EJ,
U.K.
6. Department of Applied Mathematics and Theoretical Physics,
University of Cambridge, Cambridge, CB3 0WA.
*Authors for correspondence: [email protected];
[email protected]
+ T.C. and K.G. contributed equally to this work.
Abstract: We observe polariton condensation in the yellow part
of the visible spectrum from a planar organic semiconductor
microcavity containing the molecular dye BODIPY-Br. We provide
experimental fingerprint of polariton condensation under
non-resonant optical excitation, including the non-linear
dependence of the emission intensity and wavelength blueshift with
increasing excitation density, single excitation pulse dispersion
imaging and real space interferometry. The latter two allow us to
visualise the collapse of the energy distribution and the
long-range coherence of the polariton condensate.
Bose-Einstein condensation (BEC) is an exotic state of matter,
wherein particles coalesce to a macroscopically occupied coherent
state. BECs have been observed for a broad range of systems such as
4He,[1] alkali-metal atoms,[2] magnons,[3] and polaritons.[4]
Beyond the beauty of the underlying physics describing the
fundamental properties and dynamics of BECs,[5] there is a range of
applications that utilise the coherence of their massive
wavefunctions, especially in the rapidly developing field of
quantum technologies.[6] Unlike other BECs, polariton condensates
(hosted in semiconductor slabs embedded in optical cavities) can
optically be pumped, and more importantly are interrogated through
photoluminescence. Indeed, due to finite cavity lifetimes,
polaritons decay in the form of photons that carry all information
of the corresponding polariton state (energy, momentum, spin and
phase). By appropriate choice of the crystalline semiconductor
host, polariton condensates can form even at room temperature.[7,8]
Circumventing the challenges of epitaxial growth, polariton
condensates were also realised in soft-matter organic
microcavities.[9,10] The room temperature operation and ease with
which organic polariton condensates can be “written” and “read” is
attractive for imprinting polariton lattices, a promising platform
for quantum simulators.[11]
A number of very different molecular systems have been used to
date as the semiconductor host in optical cavities to demonstrate
polariton condensation, varying from crystalline anthracene to
oligofluorenes, conjugated polymers and fluorescent
proteins.[12,9,10,13] Although the chemical structure and
morphology of these molecular hosts are very dissimilar, all three
have relatively narrow absorption spectra and small Stokes shift
with respect to their optical oscillator strengths. These features
are necessary for strong coupling, but not sufficient for polariton
condensation. For example, J-aggregates have very similar
spectroscopic characteristics with the above molecular systems
allowing for strong coupling, however, in J-aggregates polariton
condensates remains elusive even under resonant excitation
schemes.[14] A characteristic of the molecular systems that have
exhibited polariton condensation is their relatively high quantum
yield (QY), >10%, even with increasing excitation densities. It
is plausible that a key ingredient for polariton condensation in
molecular systems is the balance between radiative and
non-radiative decay paths, since the latter will limit the number
of excitons available to form polaritons. Undoubtedly, there is
still a plethora of molecular systems that satisfy the above
conditions and other considerations should be taken into account to
address limitation beyond the formation of a polariton condensate.
For any application that utilises the macroscopic wavefunction of
organic polariton condensates to become viable, control over the
disorder and its adverse effects on spatial coherence and
homogeneity of the polariton energy landscape need to be addressed.
Here, we demonstrate evidence of polariton condensation in a
different class of organic semiconductors, that of a molecular dye
diluted in a matrix polymer, resulting in extended polariton
condensates.
The material investigated is a bromine-substituted
boron-dipyrromethene (BODIPY-Br) that combines relatively high
photoluminescence quantum yield, between 10% and 50% (dependent on
sample preparation conditions), with fluorescence emission around
550 nm.[15] To incorporate BODIPY-Br into a microcavity, it is
necessary to disperse it into a transparent polymeric matrix at
relatively low concentration to limit unwanted molecular
aggregation and luminescence quenching phenomena.[16–18] This
approach to creation of a medium is different from previous studies
in which a pure film of organic chromophores was used.[9,12]
Figure 1(a) shows the chemical structure of the molecular dye
BODIPY-Br, together with its absorption and fluorescence when
dispersed into a transparent polystyrene matrix at a concentration
of 10% by mass. It can be seen that the absorption maximum of
BODIPY-Br peaks at 530 nm, whilst its fluorescence maximum is
red-shifted to 550 nm. We have previously explored the photophysics
of microcavities containing BODIPY-Br, and have concluded that
weakly-coupled excimer-like states, together with emission from the
(0,1) vibrational transition that are both located around 593 nm
are responsible for optically pumping polariton states along the
lower polariton branch.[15]
To explore the non-linear emission properties of BODIPY-Br, we
used a stripe-pumping geometry to generate amplified spontaneous
emission from a non-cavity control film. Here a 186 nm thick
non-cavity film of BODIPY-Br in polystyrene (10% by mass) was
spin-cast onto a quartz substrate and then pumped using 100 fs
laser pulses at a repetition rate of 1kHz at 500 nm along a 20 mm
long stripe. Figure 1(b) shows emission from the film at various
pump fluences. Here, it can be seen that at threshold of 104.1
μJ/cm2, there is a strong increase in emission at around 589 nm
with the emergence of a peak having a linewidth of around ~8 nm (28
meV). This wavelength approximately coincides with the (0,1)
vibrational transition, suggestive of a 4-level lasing system. This
measurement usefully indicates that BODIPY-Br can support optical
amplification, which is indicative of low optical losses under high
excitation density, rendering it a promising candidate for
realising polariton condensation.
We fabricate BODIPY-Br into a microcavity by spin-casting a
186nm thin film onto a distributed Bragg reflector (DBR) consisting
of 10 pairs of SiO2 / Nb2O5. We deposit a second 8-pair SiO2 /
Nb2O5 DBR onto the BODIPY-Br film using ion assisted electron beam
and reactive sublimation, as shown in the schematic of the cavity
structure in Figure 2(a). The measured Q-factor of the resulting
cavity is ~440 corresponding to a cavity lifetime of 133 fs. Figure
2(b) shows white light reflectivity spectra as a function of the
off-axis viewing angle, where the measurements are made through the
8-pair DBR. It can be seen that two optical modes are clearly
visible undergoing anti-crossing around a wavelength of 530 nm; a
wavelength that we associate with the (0,0) monomer absorption
transition of the BODIPY-Br. Figure 2(c) shows the energy of the
upper and lower polariton branches determined from the reflectivity
measurements. The data are fitted to a standard two-level
oscillator model from which we obtain a Rabi-splitting energy of 91
meV. The energy of the exciton energy of the fit (2.33 eV) closely
coincides with the peak absorption energy of the BODIPY-Br (0,0)
electronic transition (2.34 eV). We overlay the dispersion plot
with an intensity map of the photoluminescence emitted by the
cavity under non-resonant pumping. Here, we excite our sample using
a bandpass filter centred at 450 nm with 10 nm bandwidth to
spectrally filter a 6 ps pulsed super-continuum laser operating at
40 MHz. The excitation density for these measurements is 1.2
μJ/cm2. It can be seen that the cavity emission is most intense
around the bottom of the lower polariton branch (corresponding to
k// = 0), and reduces in intensity towards the energy of the
uncoupled exciton. We have previously shown that the distribution
of emission along the lower polariton branch (LPB) is primarily
determined by the distribution of weakly-coupled states within the
exciton reservoir.[15] Figure 2(d) shows the free space yellow
emission from our cavity under non-resonant pumping.
We investigate the non-linear photoluminescence dynamics using 2
ps optical pulses at 400 nm. The sample is held in a vacuum chamber
at 10-6mbar at room temperature. The full width at half maximum of
the pump beam on the sample is ~8 µm. Photoluminescence is
collected using a lens with 0.42 numerical aperture that allows for
dispersion and real space imaging using an electron multiplication
charge coupled device (CCD) camera at the exit slit of a 55cm
spectrophotometer and a grating of 1200 grooves/millimetre. We have
synchronised the timing of the optical detection system with the
train of excitation pulses down to a single pulse excitation/image
acquisition. Using the setup described above we can perform single
pulse excitation dispersion imaging. Also with the use of a
stabilised Michelson interferometer, we can record interferograms
that allow us to measure the extent of the coherence of the
emissive states under single pulse excitation.
Figures 3(a) - (c) show dispersion images recorded below, near
and twice above threshold. Figure 3(a) shows the linear dispersion
time averaged over 1500 excitation pulses. With increasing pump
intensity, at a threshold excitation density we observe the
collapse of the emission to the bottom of the polariton dispersion,
centred on k// = 0, as shown in the dispersion image of Fig. 3(b)
that is time averaged over 2 excitation pulses. Figure 3(c) shows
the dispersion image at approximately twice above threshold,
integrated over 4 excitation pulses, where the emission appears
blueshifted with respect to the linear regime. In Figure 3(d), we
plot the photoluminescence spectra extracted at k// = 0 from
dispersion images recorded for different pump intensities and
normalised to the number of excitation pulses used per recording.
Evidently, as the pump intensity increases the spectrum gets
narrower and blueshifts with respect to the spectrum at the lowest
excitation density (black curve in Fig. 3(d)). In Figure 3(e), we
plot the integrated intensity of the above spectra and the
corresponding linewidth measured at FWHM as a function of the upper
bound of the threshold excitation density (527.3 µJ/cm2). We
observe a threshold-like behaviour at 527.3 µJ/cm2 accompanied by a
reduction in linewidth from 2.8 to 0.5 nm. Figure 3(f) shows a
continuous blue-shift of the photoluminescence spectrum, up to 5
meV at twice the threshold density, of similar magnitude to that
observed by Plumhof et al.,[10] in a cavity containing a conjugated
polymer. In areas on the sample where we do not observe polariton
condensation, there is no visible blue-shift of the dispersion. For
the detuning used here, we calculate the exciton content, |XLP|2,
to be 0.103 at k// = 0 of the lower polariton branch. We note here
that by ramping the excitation density twice above threshold and
back to the linear regime the optical properties of the sample
remain virtually unchanged, including the shape of the
photoluminescence spectrum. The observed non-linearity on the
excitation density associated with the continuous blue-shift of the
photoluminescence spectrum provide strong evidence of polariton
condensation.
Figure 4(a) shows the interferogram of a polariton condensate at
threshold excitation density for a single excitation pulse
utilising a stabilised Michelson interferometer equipped with a
retro-reflector to generate optical interference between the
emitted light by the cavity and its inverted image. This technique
has been utilised to realise experimentally the long range phase
coherence of the condensate in many previous organic condensate
studies.[19–22] The interference fringes across the image are
plotted in Fig. 4(b) and are fitted with a convoluted Gaussian (red
curve). The visibility contrast of 86% indicates a high degree of
spatial coherence across the condensate that extends beyond the
excitation spot. From the standard deviation of the Gaussian fit we
calculate a coherence length of 28 µm from , suggesting a
homogeneous polariton energy landscape of the same order.
In conclusion, we report on a yellow polariton condensate in a
dye filled microcavity. We obtain strong evidence of non-linear
photoluminescence with increasing excitation density, associated
with a six-fold linewidth narrowing and a continuous blue-shift
attributed to polariton interactions with other polaritons and the
exciton reservoir. The wavelength of the polariton condensate is
blue-shifted with respect to the wavelength, where amplified
spontaneous emission occurs in BODIPY-Br (565 and 589 nm
respectively) indicating that the non-linearity does not
necessarily coincide with the point of maximum gain of the
molecular dye. Furthermore, single shot interferometry reveals
substantial long range coherence across the condensate and
uniformity of the polariton energy landscape. The latter is
important for expanding this work to lattices of polariton
condensates and their applications in polariton simulators. It is
conceivable that there is a large number of different molecular
dyes that could be dispersed into a polymer matrix and embedded in
a cavity that would allow for polariton condensation to be realised
at wavelengths spanning the entire visible and near infrared. This
wavelength selectivity could become of interest in the development
of future optoelectronic devices, including hybrid
organic-inorganic polariton laser diodes.
We thank EPSRC for the funding this research through the
Programme Grant EP/M025330/1 “Hybrid Polaritonics” and for funding
PhD scholarships for T.C., K.G. and R.T.G. through institutional
DTP allocations. K.G. fabricated the sample and characterised the
linear dynamics of the microcavity. T.C. contributed the non-linear
spectroscopy and A.Z. contributed the ASE measurements.
References
[1]F. Pereira Dos Santos, J. Léonard, J. Wang, C. J. Barrelet,
F. Perales, E. Rasel, C. S. Unnikrishnan, M. Leduc, C.
Cohen-Tannoudji, Phys. Rev. Lett. 2001, 86, 3459.
[2]M. H. Anderson, J. R. Ensher, M. R. Matthews, C. E. Wieman,
E. A. Cornell, Science 1995, 269, 198.
[3]V. Zapf, M. Jaime, C. D. Batista, Rev. Mod. Phys. 2014, 86,
563.
[4]J. Kasprzak, M. Richard, S. Kundermann, A. Baas, P. Jeambrun,
J. M. J. Keeling, F. M. Marchetti, M. H. Szymańska, R. André, J. L.
Staehli, V. Savona, P. B. Littlewood, B. Deveaud, L. S. Dang,
Nature 2006, 443, 409.
[5]J. R. Anglin, W. Ketterle, Nature 2002, 416, 211.
[6]K. R. Patton, U. R. Fischer, Phys. Rev. A 2013, 87,
52303.
[7]G. Christmann, R. Butté, E. Feltin, J.-F. Carlin, N.
Grandjean, Appl. Phys. Lett. 2008, 93, 51102.
[8]T.-C. Lu, Y.-Y. Lai, Y.-P. Lan, S.-W. Huang, J.-R. Chen,
Y.-C. Wu, W.-F. Hsieh, H. Deng, Opt. Express 2012, 20, 5530.
[9]K. S. Daskalakis, S. a Maier, R. Murray, S. Kéna-cohen, Nat.
Mater. 2014, 13, 271.
[10]J. D. Plumhof, T. Stoeferle, L. Mai, U. Scherf, R. F. Mahrt,
Nat. Mater. 2014, 13, 328.
[11]N. G. Berloff, K. Kalinin, M. Silva, W. Langbein, P. G.
Lagoudakis, ArXiv160706065 Cond-Mat Physicsquant-Ph 2016.
[12]S. Kéna-cohen, S. R. Forrest, Nat. Photonics 2010, DOI
10.1038/nphoton.2010.86.
[13]C. P. Dietrich, A. Steude, L. Tropf, M. Schubert, N. M.
Kronenberg, K. Ostermann, S. Höfling, M. C. Gather, Sci. Adv. 2016,
2, e1600666.
[14]N. Somaschi, L. Mouchliadis, D. Coles, I. E. Perakis, D. G.
Lidzey, P. G. Lagoudakis, P. G. Savvidis, Appl. Phys. Lett. 2011,
99, 143303.
[15]R. T. Grant, P. Michetti, A. J. Musser, P. Gregoire, T.
Virgili, E. Vella, M. Cavazzini, K. Georgiou, F. Galeotti, C.
Clark, J. Clark, C. Silva, D. G. Lidzey, Adv. Opt. Mater. 2016.
[16]T. T. Vu, M. Dvorko, E. Y. Schmidt, J. F. Audibert, P.
Retailleau, B. A. Trofimov, R. B. Pansu, G. Clavier, R.
Méallet-Renault, J. Phys. Chem. C 2013.
[17]I. Mikhalyov, N. Gretskaya, F. Bergström, L. B.-A.
Johansson, Phys. Chem. Chem. Phys. 2002, 4, 5663.
[18]S. Choi, J. Bouffard, Y. Kim, Chem. Sci. 2014, 5, 751.
[19]H. Deng, G. S. Solomon, R. Hey, K. H. Ploog, Y. Yamamoto,
Phys. Rev. Lett. 2007, 99, 126403.
[20]G. Roumpos, M. Lohse, W. H. Nitsche, J. Keeling, M. H.
Szymańska, P. B. Littlewood, A. Löffler, S. Höfling, L. Worschech,
A. Forchel, Y. Yamamoto, Proc. Natl. Acad. Sci. 2012, 109,
6467.
[21]D. Caputo, D. Ballarini, G. Dagvadorj, C. S. Muñoz, M. De
Giorgi, L. Dominici, K. West, L. N. Pfeiffer, G. Gigli, F. P.
Laussy, M. H. Szymańska, D. Sanvitto, ArXiv161005737 Cond-Mat
2016.
[22]K. S. Daskalakis, S. A. Maier, S. Kéna-Cohen, Phys. Rev.
Lett. 2015, 115, 35301.
Experimental methods
Sample preparation
A polymer matrix solution was prepared using polystyrene (PS)
having an average molecular weight (Mw) of ~192,000 in toluene at a
concentration of 35mg/ml. The PS/toluene solution was heated up to
a temperature of 70ºC and stirred for 30 minutes. BODIPY-Br was
then added to the solution at a concentration of 10% by mass.
Non-cavity films for absorption, photoluminescence and ASE
measurements were spin-cast onto quartz-coated glass
substrates.
For the microcavity fabrication a bottom 10-pair distributed
Bragg reflector (DBR) of SiO2/Nb2O5 was deposited onto
quartz-coated glass substrates using ion assisted electron beam
(Nb2O5) and thermal evaporation (SiO2). The 186nm thick BODIPY-Br
active layer was then spin-coated on top of the bottom mirror. A
second 8-pair DBR was deposited on top of the organic with the ion
gun kept turned-off during the first few layers to avoid any damage
on the organic material.
Spectroscopy
The absorption and PL measurements of the BODIPY-Br non-cavity
films were performed using a Fluoromax 4 fluorometer (Horiba) that
utilizes a Xe lamp. The angular white light reflectivity
measurement was performed using a fibre-coupled Halogen-Deuterium
white light source. A motorized arm was used to allow for the
different illumination angles between the sample and the white
light source. The reflected light was collected and coupled into an
optical fibre mounted on a second motorized arm and then sent into
a Andor Shamrock CCD spectrometer.
Angular PL measurement were performed using the same motorized
goniometer setup described earlier. A fibre-coupled Fianium
supercontinuum laser with 6ps pulses and 40MHz repetition rates was
used to non-resonantly excite the sample at 450nm using a bandpass
filter. The sample was excited at a fixed angle of 45o following a
slight downwards tilt to the optical axis to avoid collection of
the reflected excitation beam. An optical fibre on a motorized arm
was used to collect a range of different angles with a resolution
of 1o. The excitation density of the laser was kept relatively low
(1.2 μJ/cm2).
The ASE measurements were carried out using optical parametric
amplifier (Coherent OPerA SOLO) pumped by high energy Ti:sapphire
regenerative amplifier system (Coherent Libra-HE) providing up to
200 uJ per pulse at 500 nm, with 100 fs pulse duration and 1 kHz
repetition rate. Line distributed beam was produced by Thorlabs
ED1-L4100 line pattern diffuser and focused on the sample using
25.4 mm lens providing 0.173 × 20 mm vertically polarized beam on
the sample. Stimulated emission of BODIPY-Br thin film was detected
from the edge of the film, in the direction of the strip and
perpendicular to the propagation direction of the incident pump
beam using Ocean Optics QE PRO spectrometer (0.7 nm spectral
resolution). All measurements were performed at room temperature in
air.
For the condensation study, the microcavity was excited
non-resonantly at a wavelength of 400 nm using pulses at a 50 kHz
repetition rate and a pulse width < 2ps from a regenerative
amplifier (Rega 9000 pumped with a Verdi V10, Coherent) seeded by a
mode-locked picosecond Ti:Sapphire oscillator (Mira 900, coherent,
pumped by the Millennia Xs, Spectra Physics) which was
frequency-doubled through an optical parametric amplifier (OPA,
Coherent). To prevent photooxidation of the sample, two optical
choppers in a master-slave configuration with a modulation
frequency of 28 Hz were used in conjunction with an optical
shutter. This limited sample excitation to between 1 and 5 pulses
over a 30 ms exposure time. The sample was mounted in a vacuum
chamber held under a dynamic vacuum of 10-6 mbar to further reduce
photooxidation.
Photoluminescence was collected in transmission using an
apochromatic Mitutoyo 50x microscope objective with a numerical
aperture NA = 0.42 and focused into a 550 mm spectrometer (Horiba
Triax 550) coupled to an electron multiplying charged-coupled
device (CCD) with a 500 nm long-pass filter blocking the residual
light from the excitation beam. The PL was spectrally and in-plane
wavevector resolved using a 1200 grooves/mm grating and a slit
width of 100 μm at the entrance of the spectrometer. The spatial
coherence measurements were obtained by splitting the PL with a
non-polarizing beam splitter in a Michelson interferometer
configuration with the mirror on one arm replaced by a
retroreflector. The PL was then coupled into the spectrometer with
the grating at zero order to spatially resolve the PL on the
EMCCD.
17
Figure 1 (a) The normalised absorption (black) and fluorescence
spectrum (red) of BODIPY-Br dispersed in a transparent polystyrene
matrix. The chemical structure is shown in the insert. (b)
Amplified spontaneous emission from a 186nm thick film of the
BODIPY-Br dispersed in a polystyrene matrix and deposited on a
quartz substrate. A threshold is observed at a pump fluence of 104
µJ/cm2 with a peak forming at 589nm (dashed-dotted line). The
dotted line indicates the polariton emission from the microcavity,
whilst the large dashed line indicates the fluorscence emission
peak shown in (a).
Figure 2 (a) Schematic of the dye filled microcavity. (b)
Reflectivity spectra taken first at an angle of 15˚ (light orange)
and at every successive 3˚ up to 51˚ (light blue). The lower
polariton branch is observed clearly in the smaller angles (orange
to pink), whilst the anti-crossing can be observed from 42˚ (dark
purple) with the upper polariton branch becoming visible from that
point onwards. The dashed lines indicate the upper polariton
branch, exciton and lower polariton branch from left to right. (c)
Polariton dispersion in the linear regime. The dispersion of the
upper polariton (UPB) and lower polariton (LPB) branches are fitted
using a standard two level oscillator model (red lines). The peaks
from the reflectivity spectra in (b) are plotted as purple
triangles showing a good fit to the polariton branches. (d) An
image of the microcavity held in a vacuum chamber showing the
yellow emission of the polariton condensate.
Figure 3 (a)-(c) Normalised dispersions taken below threshold,
near threshold and above threshold. The white dashed line indicates
the linear regime. (d) Photoluminescence spectra extracted at k// =
0 from dispersion images recorded for different pump intensities
and normalised to the number of excitation pulses used per
recording. (e) Integrated intensity of the spectra and the
corresponding linewidth measured at full width at half maximum
(FWHM) as a function of threshold excitation density. (f) Energy
shift of the photoluminescence spectrum as a function of threshold
excitation density.
Figure 4 (a) Interferogram of a polariton condensate at
threshold excitation density for a single excitation pulse. (b)
Intensity profile (black line) taken along the white dashed line of
(a) and the corresponding Gaussian fit to the data (red curve).
1.0
0.8
0.6
0.4
0.2
0.0
Nor
mai
lsed
abs
orpt
ion
700650600550500450Wavelength (nm)
1.0
0.8
0.6
0.4
0.2
0.0
Norm
alised fluorescence
(a)
102
103
104
105
Intensity (arb. units)
680660640620600580560540Wavelength (nm)
(b) 34.7µJ/cm2
69.4µJ/cm2
104.1µJ/cm2
121.4µJ/cm2
1.0
0.8
0.6
0.4
0.2
0.0
N
o
r
m
a
i
l
s
e
d
a
b
s
o
r
p
t
i
o
n
700650600550500450
Wavelength (nm)
1.0
0.8
0.6
0.4
0.2
0.0
N
o
r
m
a
l
i
s
e
d
f
l
u
o
r
e
s
c
e
n
c
e
(a)
10
2
10
3
10
4
10
5
I
n
t
e
n
s
i
t
y
(
a
r
b
.
u
n
i
t
s
)
680660640620600580560540
Wavelength (nm)
(b)
34.7µJ/cm
2
69.4µJ/cm
2
104.1µJ/cm
2
121.4µJ/cm
2
570
565
560
555
2.242.22
2.22.18
-4 -2 0 2 4Wavevector (µm
-1)
(a)
P = 0.07Pth
570
565
560
555
-20 -10 0 10 20Angle (˚)
2.242.22
2.22.18
(c)
P = 12.3mW
P=1.9Pth
10-3
10-2
10-1
100
101
102
103
PL in
tens
ity (a
rb. u
nits
)
570568566564562560Wavelength (nm)
2.21 2.2 2.19 2.18Energy (eV)
(d) High
Low
-10
-5
0
5
10
Ener
gy sh
ift (m
eV)
2.01.51.00.5P/Pth
3
2
1
0
-1
-2
-3
Wavelength shift (nm
)
(f)
35
30
25
20
15
10
5
0Int
egra
ted
inte
nsity
(arb
. uni
ts)
2.5
2.0
1.5
1.0
0.5
Linewidth (nm
)
(e)
570
565
560
555
Wav
elen
gth
(nm
)
2.242.22
2.22.18
Energy (eV)
(b)
P = ~Pth
5
7
0
5
6
5
5
6
0
5
5
5
2
.
2
4
2
.
2
2
2
.
2
2
.
1
8
-4-2024
Wavevector (µm
-1
)
(a)
P = 0.07P
th
5
7
0
5
6
5
5
6
0
5
5
5
-20-1001020
Angle (˚)
2
.
2
4
2
.
2
2
2
.
2
2
.
1
8
(c)
P = 12.3mW
P=1.9P
th
10
-3
10
-2
10
-1
10
0
10
1
10
2
10
3
P
L
i
n
t
e
n
s
i
t
y
(
a
r
b
.
u
n
i
t
s
)
570568566564562560
Wavelength (nm)
2.212.22.192.18
Energy (eV)
(d)
High
Low
-10
-5
0
5
10
E
n
e
r
g
y
s
h
i
f
t
(
m
e
V
)
2.01.51.00.5
P/P
th
3
2
1
0
-1
-2
-3
W
a
v
e
l
e
n
g
t
h
s
h
i
f
t
(
n
m
)
(f)
35
30
25
20
15
10
5
0
I
n
t
e
g
r
a
t
e
d
i
n
t
e
n
s
i
t
y
(
a
r
b
.
u
n
i
t
s
)
2.5
2.0
1.5
1.0
0.5
L
i
n
e
w
i
d
t
h
(
n
m
)
(e)
5
7
0
5
6
5
5
6
0
5
5
5
W
a
v
e
l
e
n
g
t
h
(
n
m
)
2
.
2
4
2
.
2
2
2
.
2
2
.
1
8
E
n
e
r
g
y
(
e
V
)
(b)
P = ~P
th
1.0
0.8
0.6
0.4
0.2
0.0
Nor
mal
ised
inte
nsity
20100-10-20Real space (µm)
(b)5µm
(a)
1.0
0.8
0.6
0.4
0.2
0.0
N
o
r
m
a
l
i
s
e
d
i
n
t
e
n
s
i
t
y
20100-10-20
Real space (µm)
(b)
5µm
(a)
